repo stringclasses 1
value | github_id int64 1.27B 4.95B | github_node_id stringlengths 18 24 | number int64 8 14.3k | html_url stringlengths 49 53 | api_url stringlengths 59 63 | title stringlengths 1 402 | body stringlengths 1 62.9k ⌀ | state stringclasses 2
values | state_reason stringclasses 4
values | locked bool 2
classes | comments_count int64 0 99 | labels listlengths 0 5 | assignees listlengths 0 5 | created_at stringdate 2022-06-09 16:28:35 2026-07-22 10:32:58 | updated_at stringdate 2022-06-12 22:18:01 2026-07-22 10:32:58 | closed_at stringdate 2022-06-12 22:18:01 2026-07-21 13:17:28 ⌀ | author_association stringclasses 3
values | milestone_title stringclasses 0
values | snapshot_id stringclasses 149
values | extracted_at stringdate 2026-04-07 13:34:13 2026-07-22 10:35:16 | author_login stringlengths 3 28 | author_id int64 1.54k 296M | author_node_id stringlengths 12 20 | author_type stringclasses 3
values | author_site_admin bool 1
class |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
huggingface/diffusers | 1,575,720,208 | I_kwDOHa8MBc5d65UQ | 2,284 | https://github.com/huggingface/diffusers/issues/2284 | https://api.github.com/repos/huggingface/diffusers/issues/2284 | stable diffusion machine memory leak | I have encountered a memory leak problem, we need to store and switch the model of stable diffusion in memory, but when we convert the model in memory to gpu, the memory is not reclaimed, and at the same time, when the model in memory is deleted, the memory also not recycled.
Therefore, what should I do to reclaim the... | closed | completed | false | 4 | [
"stale"
] | [] | 2023-02-08T08:49:14Z | 2023-04-18T13:15:44Z | 2023-03-19T15:03:01Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | wen020 | 54,690,997 | MDQ6VXNlcjU0NjkwOTk3 | User | false |
huggingface/diffusers | 1,576,025,906 | I_kwDOHa8MBc5d8D8y | 2,288 | https://github.com/huggingface/diffusers/issues/2288 | https://api.github.com/repos/huggingface/diffusers/issues/2288 | [Examples] Implement a training script for InstructPix2Pix | We already have the pipeline: https://github.com/huggingface/diffusers/pull/2040
Given how interesting results it can generate, I think it would be greatly beneficial to add an example script showing how one can perform the underlying training of InstructPix2Pix -- another useful way to fine-tune Stable Diffusion :)... | closed | completed | false | 2 | [
"stale"
] | [
"sayakpaul"
] | 2023-02-08T12:24:09Z | 2023-03-23T04:45:03Z | 2023-03-23T04:45:03Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | sayakpaul | 22,957,388 | MDQ6VXNlcjIyOTU3Mzg4 | User | false |
huggingface/diffusers | 1,346,131,127 | I_kwDOHa8MBc5QPFS3 | 229 | https://github.com/huggingface/diffusers/issues/229 | https://api.github.com/repos/huggingface/diffusers/issues/229 | "Safety Module" will be used to *create* NSFW images | ### Describe the bug
You have added a "Safety Module" that returns black images for NSFW content.
https://github.com/huggingface/diffusers/commit/65ea7d6b628d9b4c0fedf59bb59e7e57c88a44ff
It is obvious that this classifier will be used for a different purpose. It is trivial to change the module so that Stable D... | closed | completed | false | 4 | [
"bug"
] | [] | 2022-08-22T09:53:47Z | 2022-08-24T06:39:33Z | 2022-08-24T06:39:33Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | rolux | 152,646 | MDQ6VXNlcjE1MjY0Ng== | User | false |
huggingface/diffusers | 1,576,557,955 | I_kwDOHa8MBc5d-F2D | 2,292 | https://github.com/huggingface/diffusers/issues/2292 | https://api.github.com/repos/huggingface/diffusers/issues/2292 | Expose transformer depth to unet config | **Is your feature request related to a problem? Please describe.**
Transformer depths is configurable but is currently hard coded to 1 when used through the unet see https://github.com/huggingface/diffusers/blob/fd5c3c09afe29cbbde42cfa791ca598a07d94da9/src/diffusers/models/unet_2d_blocks.py#L775
**Describe the ... | closed | completed | false | 1 | [
"stale"
] | [
"williamberman"
] | 2023-02-08T17:53:58Z | 2023-03-30T15:03:39Z | 2023-03-30T15:03:39Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | williamberman | 6,505,431 | MDQ6VXNlcjY1MDU0MzE= | User | false |
huggingface/diffusers | 1,576,629,271 | I_kwDOHa8MBc5d-XQX | 2,294 | https://github.com/huggingface/diffusers/issues/2294 | https://api.github.com/repos/huggingface/diffusers/issues/2294 | ImportError: cannot import name 'StableDiffusionVideoInpaintPipeline' from 'diffusers' | ### Describe the bug
I installed diffusers using pip install --upgrade diffusers[torch] and got returned the following error.
ImportError: cannot import name 'StableDiffusionVideoInpaintPipeline' from 'diffusers'
I get the same return when installing with conda.
Any idea how to rememdy this issue?
### Reprod... | closed | completed | false | 2 | [
"bug",
"stale"
] | [] | 2023-02-08T18:50:39Z | 2023-03-19T15:02:58Z | 2023-03-19T15:02:58Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | reillydonovan | 5,108,592 | MDQ6VXNlcjUxMDg1OTI= | User | false |
huggingface/diffusers | 1,576,810,057 | I_kwDOHa8MBc5d_DZJ | 2,296 | https://github.com/huggingface/diffusers/issues/2296 | https://api.github.com/repos/huggingface/diffusers/issues/2296 | enable_xformers_memory_efficient_attention not work in T4 GPU for stable-diffusion-v1-5 | ### Describe the bug
Use enable_xformers_memory_efficient_attention get the following error in T4 GPU:
```
ValueError: Operator `memory_efficient_attention` does not support inputs:
query : shape=(16, 256, 1, 160) (torch.float16)
key : shape=(16, 256, 1, 160) (torch.float16)
value ... | closed | completed | false | 2 | [
"bug"
] | [] | 2023-02-08T21:16:12Z | 2023-02-19T17:32:38Z | 2023-02-19T17:32:38Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | tianleiwu | 30,328,909 | MDQ6VXNlcjMwMzI4OTA5 | User | false |
huggingface/diffusers | 1,577,484,364 | I_kwDOHa8MBc5eBoBM | 2,299 | https://github.com/huggingface/diffusers/issues/2299 | https://api.github.com/repos/huggingface/diffusers/issues/2299 | fast-stable-diffusion fine-tuned checkpoints in Diffusers | ### Describe the bug
Hello,
I wanted to bring up the dreambooth model through the following command.
```
pipe = StableDiffusionPipeline.from_pretrained("./test.ckpt")
```
Valid Error "model_index.json"
The address of the model I learned is fast-dreamboot and I used it at the following address.
"https://git... | closed | completed | false | 2 | [
"bug"
] | [] | 2023-02-09T08:51:47Z | 2023-02-09T14:22:43Z | 2023-02-09T14:22:42Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | kimtaehyeong | 11,425,786 | MDQ6VXNlcjExNDI1Nzg2 | User | false |
huggingface/diffusers | 1,275,402,881 | I_kwDOHa8MBc5MBRqB | 23 | https://github.com/huggingface/diffusers/issues/23 | https://api.github.com/repos/huggingface/diffusers/issues/23 | Adding Diffusion Planning Model from "Planning with Diffusion for Flexible Behavior Synthesis" | Hi! First off, thanks for creating this awesome library, excited to see where it goes!
I did see in the README that there's a TODO for adding the RL model from https://github.com/jannerm/diffuser, and I recently saw https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/unet_rl.py. Is this actively... | closed | completed | false | 0 | [] | [] | 2022-06-17T19:51:46Z | 2022-06-20T05:05:32Z | 2022-06-20T05:05:32Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | shyamsn97 | 18,158,178 | MDQ6VXNlcjE4MTU4MTc4 | User | false |
huggingface/diffusers | 1,577,510,695 | I_kwDOHa8MBc5eBucn | 2,300 | https://github.com/huggingface/diffusers/issues/2300 | https://api.github.com/repos/huggingface/diffusers/issues/2300 | Is there support to convert the `bf16` models from `flax` to `onnx` | Currently the `bf-16` models in https://huggingface.co/runwayml/stable-diffusion-v1-5/tree/bf16/unet & https://huggingface.co/CompVis/stable-diffusion-v1-4/tree/bf16 are all flax models | closed | completed | false | 8 | [
"stale"
] | [] | 2023-02-09T09:09:40Z | 2023-03-19T15:02:57Z | 2023-03-19T15:02:57Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | lalith-mcw | 84,698,958 | MDQ6VXNlcjg0Njk4OTU4 | User | false |
huggingface/diffusers | 1,577,604,575 | I_kwDOHa8MBc5eCFXf | 2,301 | https://github.com/huggingface/diffusers/issues/2301 | https://api.github.com/repos/huggingface/diffusers/issues/2301 | Question about alpha in imagic_stable_diffusion | Hi, thanks for the excellent work. I have a question about alpha in [imagic](https://github.com/huggingface/diffusers/blob/a7ca03aa85f94574f06576d2155b3ec061fe8d63/examples/community/imagic_stable_diffusion.py#L399). In imgaic paper, the interpolation ratio between $e_{tgt}$ and $e_{opt}$ is (0, 1). However, the ratio ... | closed | completed | false | 4 | [] | [] | 2023-02-09T10:13:27Z | 2023-02-12T14:28:01Z | 2023-02-12T07:08:30Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | cyysc1998 | 35,159,700 | MDQ6VXNlcjM1MTU5NzAw | User | false |
huggingface/diffusers | 1,577,733,782 | I_kwDOHa8MBc5eCk6W | 2,304 | https://github.com/huggingface/diffusers/issues/2304 | https://api.github.com/repos/huggingface/diffusers/issues/2304 | Create VAE feature extractor class | We currently do not have a unified API that makes sure that for pipelines that accept image inputs that image input and output format always stay the same. This PR shows that nicely: https://github.com/huggingface/diffusers/issues/1882#issuecomment-1416117217
We should make sure that:
- a) All pipelines that acce... | closed | completed | false | 9 | [
"stale"
] | [
"yiyixuxu"
] | 2023-02-09T11:40:47Z | 2023-04-04T17:40:27Z | 2023-04-03T14:46:40Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | patrickvonplaten | 23,423,619 | MDQ6VXNlcjIzNDIzNjE5 | User | false |
huggingface/diffusers | 1,578,054,216 | I_kwDOHa8MBc5eDzJI | 2,306 | https://github.com/huggingface/diffusers/issues/2306 | https://api.github.com/repos/huggingface/diffusers/issues/2306 | INT_MAX error | ### Describe the bug
Running the Stable Diffusion 2 generation pipeline with fp16, attention slicing and batch size 16 outputs error message: "RuntimeError: upsample_nearest_nhwc only supports output tensors with less than INT_MAX elements". EDIT: I can go up to batch size 14, but for >=15 I receive this error.
#... | closed | completed | false | 3 | [
"bug",
"stale"
] | [] | 2023-02-09T15:00:11Z | 2023-03-19T15:02:56Z | 2023-03-19T15:02:56Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | morrisalp | 8,263,996 | MDQ6VXNlcjgyNjM5OTY= | User | false |
huggingface/diffusers | 1,578,210,682 | I_kwDOHa8MBc5eEZV6 | 2,307 | https://github.com/huggingface/diffusers/issues/2307 | https://api.github.com/repos/huggingface/diffusers/issues/2307 | StableDiffusion pipelines - negative prompt | Hi,
Maybe I misunderstand something but in most of the stable diffusion pipelines after we embed negative prompt - there is a concatenation of `negative_prompt_embedding` with `prompt_embedding`:
`prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])`
Shouldn't it be rather something like:
`pro... | closed | completed | false | 4 | [
"stale"
] | [] | 2023-02-09T16:26:08Z | 2023-03-20T15:03:24Z | 2023-03-20T15:03:24Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | kdziedzic66 | 12,718,383 | MDQ6VXNlcjEyNzE4Mzgz | User | false |
huggingface/diffusers | 1,578,779,945 | I_kwDOHa8MBc5eGkUp | 2,309 | https://github.com/huggingface/diffusers/issues/2309 | https://api.github.com/repos/huggingface/diffusers/issues/2309 | Unable to run textual inversion example in AWS Sagemaker | **Description**
I am trying to run the textual inversion example in AWS Sagemaker, but I keep getting the following error:
TypeError: SageMakerConfig.__init__() missing 1 required positional argument: 'image_uri'
I have followed all of the instructions listed in the readME and this is what my accelerate_cofig.ya... | closed | completed | false | 2 | [
"stale"
] | [] | 2023-02-09T23:51:24Z | 2023-03-20T15:03:22Z | 2023-03-20T15:03:22Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | karleikongsiri | 82,921,322 | MDQ6VXNlcjgyOTIxMzIy | User | false |
huggingface/diffusers | 1,578,854,659 | I_kwDOHa8MBc5eG2kD | 2,310 | https://github.com/huggingface/diffusers/issues/2310 | https://api.github.com/repos/huggingface/diffusers/issues/2310 | LDM Inpainting, LaMA cleaner | ### Model/Pipeline/Scheduler description
Hi,
Looking to see if we have any inputs on getting the LDM with inpaint, as that is advanced today with cleaning of image given a mask kind of. Similar to LaMA, ZITS etc.
It would be also great if we could get LaMA, ZITS etc pipelines as well.
### Open source status
... | closed | completed | false | 3 | [
"stale"
] | [] | 2023-02-10T01:36:55Z | 2023-03-20T15:03:21Z | 2023-03-20T15:03:20Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | polavishnu4444 | 2,263,240 | MDQ6VXNlcjIyNjMyNDA= | User | false |
huggingface/diffusers | 1,579,090,279 | I_kwDOHa8MBc5eHwFn | 2,311 | https://github.com/huggingface/diffusers/issues/2311 | https://api.github.com/repos/huggingface/diffusers/issues/2311 | UniPC Scheduler | ### Model/Pipeline/Scheduler description
New fast scheduler:
https://github.com/wl-zhao/UniPC/tree/main
### Open source status
- [X] The model implementation is available
- [X] The model weights are available (Only relevant if addition is not a scheduler).
### Provide useful links for the implementation
@wl-zhao | closed | completed | false | 6 | [
"New scheduler"
] | [] | 2023-02-10T06:51:17Z | 2023-02-20T13:00:30Z | 2023-02-20T13:00:30Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | patrickvonplaten | 23,423,619 | MDQ6VXNlcjIzNDIzNjE5 | User | false |
huggingface/diffusers | 1,579,135,127 | I_kwDOHa8MBc5eH7CX | 2,312 | https://github.com/huggingface/diffusers/issues/2312 | https://api.github.com/repos/huggingface/diffusers/issues/2312 | Text to Image Finetune Save EMA hook throws error | ### Describe the bug
I think this commit `f5ccffecf7c9c7b08c39043d1c9528be2c5cfecc` on Feb 7 2023 has introduced a bug where in this function:
``` def save_model_hook(models, weights, output_dir):
if args.use_ema:
ema_unet.save_pretrained(os.path.join(output_dir, "unet_ema"))
... | closed | completed | false | 4 | [
"bug"
] | [
"williamberman",
"patrickvonplaten"
] | 2023-02-10T07:24:36Z | 2023-02-16T13:40:30Z | 2023-02-16T13:40:30Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | samar-khanna | 39,067,298 | MDQ6VXNlcjM5MDY3Mjk4 | User | false |
huggingface/diffusers | 1,580,226,765 | I_kwDOHa8MBc5eMFjN | 2,315 | https://github.com/huggingface/diffusers/issues/2315 | https://api.github.com/repos/huggingface/diffusers/issues/2315 | scheduler_config.json | I notice when I train a custom model with train_deambooth.py there are two different schedulers, a regular scheduler and a noise scheduler, but there is only one scheduler_config.json in my directory when finished. In this case it is DDIM and DDPM for the noise scheduler. Both of these schedulers have different beta_... | closed | completed | false | 5 | [] | [
"williamberman",
"patrickvonplaten"
] | 2023-02-10T19:40:32Z | 2023-03-06T21:02:46Z | 2023-03-06T21:02:46Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | rmac85 | 90,567,763 | MDQ6VXNlcjkwNTY3NzYz | User | false |
huggingface/diffusers | 1,580,767,145 | I_kwDOHa8MBc5eOJep | 2,319 | https://github.com/huggingface/diffusers/issues/2319 | https://api.github.com/repos/huggingface/diffusers/issues/2319 | Typo in load_pipeline_from_original_stable_diffusion_ckpt() method | ### Describe the bug
Currently it works fine, but I think `inferenc.yaml` should be `inference.yaml`.
https://github.com/huggingface/diffusers/blob/1e7f96544203724652707dc93a7671c90efd3eeb/src/diffusers/pipelines/stable_diffusion/convert_from_ckpt.py#L870-L894
### Reproduction
_No response_
### Logs
_No respons... | closed | completed | false | 1 | [
"bug"
] | [] | 2023-02-11T09:58:27Z | 2023-02-19T18:03:16Z | 2023-02-19T18:03:16Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | p1atdev | 60,182,057 | MDQ6VXNlcjYwMTgyMDU3 | User | false |
huggingface/diffusers | 1,346,962,504 | I_kwDOHa8MBc5QSQRI | 232 | https://github.com/huggingface/diffusers/issues/232 | https://api.github.com/repos/huggingface/diffusers/issues/232 | Working offline once the models have been downloaded | I thought that once the models had been downloaded once, they would be cached and it would be possible to use it offline.
Using [this code](https://github.com/huggingface/diffusers/issues/228#issuecomment-1222813519), you need to have internet working at the beginning of the program.
Once it runs, internet is no... | closed | completed | false | 2 | [] | [] | 2022-08-22T20:44:07Z | 2022-08-22T21:26:37Z | 2022-08-22T21:26:37Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | leszekhanusz | 4,641,499 | MDQ6VXNlcjQ2NDE0OTk= | User | false |
huggingface/diffusers | 1,581,074,574 | I_kwDOHa8MBc5ePUiO | 2,321 | https://github.com/huggingface/diffusers/issues/2321 | https://api.github.com/repos/huggingface/diffusers/issues/2321 | Not copying the model in inference and assigning ema_weights seems incorrect. | ### Describe the bug
This recent commit https://github.com/huggingface/diffusers/commit/44f6bc81c764897406952bf75263b866de7c8cd3 regarding the PR #2166 removed copying the model while doing inference. Is this correct ? Aren't we substituting EMA weight to the underlying model (in the next line) which is being traine... | closed | completed | false | 8 | [
"bug",
"stale"
] | [
"williamberman",
"patil-suraj"
] | 2023-02-12T01:06:53Z | 2023-03-16T23:16:01Z | 2023-03-16T23:16:01Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | dasayan05 | 18,719,603 | MDQ6VXNlcjE4NzE5NjAz | User | false |
huggingface/diffusers | 1,581,286,623 | I_kwDOHa8MBc5eQITf | 2,323 | https://github.com/huggingface/diffusers/issues/2323 | https://api.github.com/repos/huggingface/diffusers/issues/2323 | Textual inversion captioning | In the examples, are we able to provide captions for textual inversion trainings or are we limited to the placeholder and initializer tokens? If I put `<image-name>.txt` in the training folder are they used? If not, is there any guidance or documentation on how to extend it? | closed | completed | false | 7 | [] | [
"williamberman",
"patrickvonplaten"
] | 2023-02-12T14:48:26Z | 2023-02-21T18:34:44Z | 2023-02-13T23:28:52Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | jndietz | 4,515,791 | MDQ6VXNlcjQ1MTU3OTE= | User | false |
huggingface/diffusers | 1,581,367,839 | I_kwDOHa8MBc5eQcIf | 2,324 | https://github.com/huggingface/diffusers/issues/2324 | https://api.github.com/repos/huggingface/diffusers/issues/2324 | Dreambooth Training not reading instance data | I am trying to run the Dreambooth Training on my local device from the line command.
However, this gave me a hard time due to the trained weights not generating images like the instance data (pictures of my face). I was trying to troubleshoot this and figured out that the generated images (from the trained weights) ... | closed | completed | false | 7 | [
"bug"
] | [
"williamberman",
"patrickvonplaten"
] | 2023-02-12T18:47:50Z | 2023-02-17T22:33:28Z | 2023-02-17T22:33:28Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | asahi0130 | 106,039,754 | U_kgDOBlIJyg | User | false |
huggingface/diffusers | 1,581,534,470 | I_kwDOHa8MBc5eRE0G | 2,326 | https://github.com/huggingface/diffusers/issues/2326 | https://api.github.com/repos/huggingface/diffusers/issues/2326 | Error converting LoRA to safetensors/ckpt | Generally speaking, I am not clear on what to do with the output of these LoRA python scripts. I don't think the output can be natively used by the webuis. Other LoRAs I've seen online are usually safetensors. Here is what I did...
I used the provided python script in `examples` to generate a LoRA:
```
#!/bin... | closed | completed | false | 43 | [] | [
"williamberman",
"patrickvonplaten"
] | 2023-02-13T02:27:39Z | 2023-08-11T20:39:56Z | 2023-02-27T05:36:41Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | jndietz | 4,515,791 | MDQ6VXNlcjQ1MTU3OTE= | User | false |
huggingface/diffusers | 1,581,722,233 | I_kwDOHa8MBc5eRyp5 | 2,328 | https://github.com/huggingface/diffusers/issues/2328 | https://api.github.com/repos/huggingface/diffusers/issues/2328 | [Pipelines] On adding Pix2Pix Zero | [Pix2Pix Zero](https://pix2pixzero.github.io/) presents a wonderful idea for editing the objects in a generated image with finer controls. Some really amazing results:
![image](https://user-images.githu... | closed | completed | false | 8 | [
"stale"
] | [
"sayakpaul"
] | 2023-02-13T06:07:05Z | 2023-03-15T15:08:56Z | 2023-03-15T15:08:56Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | sayakpaul | 22,957,388 | MDQ6VXNlcjIyOTU3Mzg4 | User | false |
huggingface/diffusers | 1,347,095,939 | I_kwDOHa8MBc5QSw2D | 233 | https://github.com/huggingface/diffusers/issues/233 | https://api.github.com/repos/huggingface/diffusers/issues/233 | How do you verify that integration is equivalent to original Stable Diffusion? | Hi there HF folks!
Potentially silly question, but I went through the [commit history](https://github.com/huggingface/diffusers/commits/71ba8aec55b52a7ba5a1ff1db1265ffdd3c65ea2/src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py) behind the stable diffusion integration and I'm wondering: how do you... | closed | completed | false | 7 | [] | [] | 2022-08-22T23:04:35Z | 2022-09-13T15:54:38Z | 2022-09-13T15:54:38Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | gordicaleksa | 29,271,842 | MDQ6VXNlcjI5MjcxODQy | User | false |
huggingface/diffusers | 1,581,983,329 | I_kwDOHa8MBc5eSyZh | 2,331 | https://github.com/huggingface/diffusers/issues/2331 | https://api.github.com/repos/huggingface/diffusers/issues/2331 | [Pipelines] Add a ControlNet pipeline | ### Model/Pipeline/Scheduler description
[ControlNet](https://github.com/lllyasviel/ControlNet) by @lllyasviel is a neural network structure to control diffusion models by adding extra conditions.
It has integration with Stable Diffusion and [8 pre-trained models](https://huggingface.co/lllyasviel/ControlNet/tr... | closed | completed | false | 47 | [
"good first issue",
"New pipeline/model"
] | [] | 2023-02-13T09:37:21Z | 2024-03-18T07:56:30Z | 2023-03-06T23:25:53Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | apolinario | 788,417 | MDQ6VXNlcjc4ODQxNw== | User | false |
huggingface/diffusers | 1,582,128,869 | I_kwDOHa8MBc5eTV7l | 2,332 | https://github.com/huggingface/diffusers/issues/2332 | https://api.github.com/repos/huggingface/diffusers/issues/2332 | When I want to run the inference after training dreambooth, I report an error | ### Model/Pipeline/Scheduler description
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda:3")
prompt = "A photo of a beautiful Asian sks girl"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7.5).images[0]
image.save("dog-bucket.png")
### Open source stat... | closed | completed | false | 4 | [] | [
"williamberman",
"patrickvonplaten"
] | 2023-02-13T11:00:07Z | 2023-02-13T22:17:24Z | 2023-02-13T22:17:23Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | 1235555555shan | 85,985,529 | MDQ6VXNlcjg1OTg1NTI5 | User | false |
huggingface/diffusers | 1,347,140,794 | I_kwDOHa8MBc5QS7y6 | 234 | https://github.com/huggingface/diffusers/issues/234 | https://api.github.com/repos/huggingface/diffusers/issues/234 | Stable Diffusion pipeline doesn't run with half-precision | ### Describe the bug
As of the 0.2.4 release, the StableDiffusion pipeline doesn't seem to work with the half-precision "fp16" weights and `torch_dtype=torch.float16`. I think it's due to [this line](https://github.com/huggingface/diffusers/blob/051b34635fda2fc310898a6a602c89be8663b77f/src/diffusers/models/embedding... | closed | completed | false | 4 | [
"bug"
] | [] | 2022-08-23T00:01:55Z | 2022-08-23T03:08:38Z | 2022-08-23T03:08:37Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | singularperturbation | 2,097,138 | MDQ6VXNlcjIwOTcxMzg= | User | false |
huggingface/diffusers | 1,583,069,262 | I_kwDOHa8MBc5eW7hO | 2,340 | https://github.com/huggingface/diffusers/issues/2340 | https://api.github.com/repos/huggingface/diffusers/issues/2340 | [testing] tiny stable diffusion pipe throws error when used with safety checker | ### Describe the bug
The tiny stable diffusion pipeline we use for some tests throws an error when used with the safety checker. It currently doesn't throw an error in any tests because we always use it without the safety checker.
https://huggingface.co/hf-internal-testing/tiny-stable-diffusion-pipe
### Reproducti... | closed | completed | false | 3 | [
"bug",
"stale"
] | [
"williamberman"
] | 2023-02-13T21:11:04Z | 2023-03-24T15:03:32Z | 2023-03-24T15:03:32Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | williamberman | 6,505,431 | MDQ6VXNlcjY1MDU0MzE= | User | false |
huggingface/diffusers | 1,583,187,235 | I_kwDOHa8MBc5eXYUj | 2,342 | https://github.com/huggingface/diffusers/issues/2342 | https://api.github.com/repos/huggingface/diffusers/issues/2342 | Possible to use vit-B/32 instead of vit-L/14 for SD fine-tuning on Dreambooth? | I am wondering if I can change the default clip model to run my training and if so, how? | closed | completed | false | 4 | [
"stale"
] | [
"yiyixuxu",
"patrickvonplaten"
] | 2023-02-13T22:47:19Z | 2023-03-24T15:03:31Z | 2023-03-24T15:03:30Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | stpg06 | 96,386,710 | U_kgDOBb6-lg | User | false |
huggingface/diffusers | 1,583,292,333 | I_kwDOHa8MBc5eXx-t | 2,343 | https://github.com/huggingface/diffusers/issues/2343 | https://api.github.com/repos/huggingface/diffusers/issues/2343 | callback_steps is typed as Optional[int] but has a default value | In pipelines, we pass both a callback and a callback_steps. callback_steps is always set and has a default value of 1. callback is optional. We always type callback_steps incorrectly as Optional[int] instead of just int. We should update the type signatures of callback_steps.
https://github.com/huggingface/diffusers... | closed | completed | false | 1 | [] | [
"williamberman"
] | 2023-02-14T00:36:41Z | 2023-02-14T22:35:31Z | 2023-02-14T22:35:31Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | williamberman | 6,505,431 | MDQ6VXNlcjY1MDU0MzE= | User | false |
huggingface/diffusers | 1,583,414,846 | I_kwDOHa8MBc5eYP4- | 2,344 | https://github.com/huggingface/diffusers/issues/2344 | https://api.github.com/repos/huggingface/diffusers/issues/2344 | Unable to run inference with provided 🤗 example scripts | This is tangentially related to my issue #2326. I believe I have successfully generated a LoRA, and would like to use it in a `DiffusionPipeline` for inference. Full disclosure: I am trying to run this on a 1050ti with 4GB VRAM. I am able to run inference from within the automatic1111 webui. Following some of the d... | closed | completed | false | 7 | [] | [] | 2023-02-14T03:11:17Z | 2023-02-16T09:56:33Z | 2023-02-14T23:14:19Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | jndietz | 4,515,791 | MDQ6VXNlcjQ1MTU3OTE= | User | false |
huggingface/diffusers | 1,583,725,623 | I_kwDOHa8MBc5eZbw3 | 2,348 | https://github.com/huggingface/diffusers/issues/2348 | https://api.github.com/repos/huggingface/diffusers/issues/2348 | Documentation for `KarrasDiffusionSchedulers` | I assume there's a general classification for the types of schedulers that can be added to this type. It would be helpful to add a quick blurb about why certain schedulers qualify as `KarrasDiffusionSchedulers` | closed | completed | false | 3 | [] | [
"kashif",
"williamberman"
] | 2023-02-14T08:33:46Z | 2023-02-15T18:20:15Z | 2023-02-15T18:20:15Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | williamberman | 6,505,431 | MDQ6VXNlcjY1MDU0MzE= | User | false |
huggingface/diffusers | 1,347,192,279 | I_kwDOHa8MBc5QTIXX | 235 | https://github.com/huggingface/diffusers/issues/235 | https://api.github.com/repos/huggingface/diffusers/issues/235 | trouble with cuda when porting colab script to standalone python | ### Describe the bug
I'm trying to write a minimal standalone python script based from the colab code.
After correctly installing all packages and setting up my huggingface token it fails upon run time. Pretty sure I haven't left anything out and the exact same code verbatim runs in colab concurrently.
### Reproduc... | closed | completed | false | 2 | [
"bug"
] | [] | 2022-08-23T01:28:26Z | 2022-09-13T15:54:53Z | 2022-09-13T15:54:53Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | TheOneTrueGuy | 3,240,071 | MDQ6VXNlcjMyNDAwNzE= | User | false |
huggingface/diffusers | 1,584,163,644 | I_kwDOHa8MBc5ebGs8 | 2,351 | https://github.com/huggingface/diffusers/issues/2351 | https://api.github.com/repos/huggingface/diffusers/issues/2351 | DEISMultistepScheduler missing get_velocity function | ### Describe the bug
scheduling_deis_multistep is almost a drop in replacement for scheduling_ddpm however it lacks the get_velocity function from scheduling_ddpm. In the project I am working on the lack of this function prevents using models that have v_prediction. The implementation from scheduling_ddpm (as below) h... | closed | completed | false | 3 | [
"bug",
"stale"
] | [
"yiyixuxu"
] | 2023-02-14T13:23:34Z | 2023-04-18T15:03:49Z | 2023-04-18T15:03:49Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | saunderez | 7,011,049 | MDQ6VXNlcjcwMTEwNDk= | User | false |
huggingface/diffusers | 1,584,545,255 | I_kwDOHa8MBc5ecj3n | 2,353 | https://github.com/huggingface/diffusers/issues/2353 | https://api.github.com/repos/huggingface/diffusers/issues/2353 | OnnxStableDiffusionInpaintPipeline lacks denoise support | `OnnxStableDiffusionInpaintPipelineLegacy` has support for denoise strength, but `OnnxStableDiffusionInpaintPipeline` does not. | closed | completed | false | 2 | [] | [] | 2023-02-14T17:18:38Z | 2023-02-16T09:53:23Z | 2023-02-16T09:53:22Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | xzuyn | 16,216,325 | MDQ6VXNlcjE2MjE2MzI1 | User | false |
huggingface/diffusers | 1,584,635,446 | I_kwDOHa8MBc5ec542 | 2,354 | https://github.com/huggingface/diffusers/issues/2354 | https://api.github.com/repos/huggingface/diffusers/issues/2354 | Has anyone successfully used pre-trained models from guided diffusion with diffusers' pipeline and scheduler? | The models in diffusers outputs sample directly, while models in guided diffusion outputs mean and variance.
Relative issue https://github.com/openai/guided-diffusion/issues/93#issue-1584632267 | closed | completed | false | 2 | [
"stale"
] | [] | 2023-02-14T18:25:35Z | 2023-03-25T15:04:08Z | 2023-03-25T15:04:08Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | ntlm1686 | 78,901,505 | MDQ6VXNlcjc4OTAxNTA1 | User | false |
huggingface/diffusers | 1,584,855,103 | I_kwDOHa8MBc5edvg_ | 2,356 | https://github.com/huggingface/diffusers/issues/2356 | https://api.github.com/repos/huggingface/diffusers/issues/2356 | [Test] Add tests that no imports have been added that are not a hard dependency | As seen in this PR: https://github.com/huggingface/diffusers/pull/2193 our tests can all pass even if dependencies are added to PRs that are no hard dependencies. The reason for this is that we use `torchvision` for example testing and thus have it in our docker container, but it's not a hard dependency and therefore a... | closed | completed | false | 6 | [
"wip"
] | [
"sayakpaul",
"patrickvonplaten"
] | 2023-02-14T21:27:22Z | 2023-06-28T16:35:27Z | 2023-06-28T16:35:27Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | patrickvonplaten | 23,423,619 | MDQ6VXNlcjIzNDIzNjE5 | User | false |
huggingface/diffusers | 1,584,936,988 | I_kwDOHa8MBc5eeDgc | 2,358 | https://github.com/huggingface/diffusers/issues/2358 | https://api.github.com/repos/huggingface/diffusers/issues/2358 | Consolidate `torch.backends.cuda.matmul.allow_tf32 = False` in tests | The module with `PipelineTesterMixin` already sets this. We shouldn't repeatedly set this in all tests and instead consolidate where we set it | closed | completed | false | 1 | [
"stale"
] | [
"williamberman"
] | 2023-02-14T22:42:51Z | 2023-03-25T15:04:05Z | 2023-03-25T15:04:05Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | williamberman | 6,505,431 | MDQ6VXNlcjY1MDU0MzE= | User | false |
huggingface/diffusers | 1,585,021,054 | I_kwDOHa8MBc5eeYB- | 2,359 | https://github.com/huggingface/diffusers/issues/2359 | https://api.github.com/repos/huggingface/diffusers/issues/2359 | Allow latent output_type when using StableDiffusionImg2ImgPipeline | **Is your feature request related to a problem? Please describe.**
Currently, the StableDiffusionImg2ImgPipeline cannot output latents the way that the base StableDiffusionPipeline can. I think this would be a very desirable feature, especially with the arrival of the latent upscaler: https://huggingface.co/stability... | closed | completed | false | 6 | [] | [
"yiyixuxu"
] | 2023-02-15T00:24:27Z | 2023-03-21T02:35:58Z | 2023-03-21T02:35:43Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | samnappi | 60,114,663 | MDQ6VXNlcjYwMTE0NjYz | User | false |
huggingface/diffusers | 1,585,440,692 | I_kwDOHa8MBc5ef-e0 | 2,360 | https://github.com/huggingface/diffusers/issues/2360 | https://api.github.com/repos/huggingface/diffusers/issues/2360 | Is there a training script for instruct pix2pix ? | Thanks for your great job. I found a test script for instruct pix2pix , but I cannot find the traning one. Where can I find it. Or do you plan to give that? thank you. | closed | completed | false | 6 | [
"stale"
] | [] | 2023-02-15T08:39:48Z | 2023-04-20T15:03:41Z | 2023-04-20T15:03:41Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | tengshaofeng | 9,246,739 | MDQ6VXNlcjkyNDY3Mzk= | User | false |
huggingface/diffusers | 1,585,476,983 | I_kwDOHa8MBc5egHV3 | 2,361 | https://github.com/huggingface/diffusers/issues/2361 | https://api.github.com/repos/huggingface/diffusers/issues/2361 | Text_to_Image finetuning script crashes when trying to save checkpoints | ### Describe the bug
Hello everyone,
I want to signal a problem I have found while fine tuning a text_to_image model.
I am a newbie so I might be missing something big.
I have also found a possible workaround to solve it.
I am currently using the guide at this link:
https://huggingface.co/docs/diffusers/trainin... | closed | completed | false | 3 | [
"bug"
] | [] | 2023-02-15T09:04:57Z | 2023-02-16T14:45:46Z | 2023-02-16T14:45:45Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | gianlucaasti | 41,920,206 | MDQ6VXNlcjQxOTIwMjA2 | User | false |
huggingface/diffusers | 1,585,664,851 | I_kwDOHa8MBc5eg1NT | 2,363 | https://github.com/huggingface/diffusers/issues/2363 | https://api.github.com/repos/huggingface/diffusers/issues/2363 | How could I convert a LoRA .safetensors or .ckpt file into the format that diffusers can process? | ### Describe the bug
I got some LoRA model in .safetensors format, and tried to convert in to the format that can be used in diffusers.
But nowhere I can find any document or scripts to achieve that.
So I try to convert the file with the [convert_original_stable_diffusion_to_diffusers.py](https://github.com/huggin... | closed | completed | false | 32 | [
"bug",
"stale"
] | [] | 2023-02-15T11:09:04Z | 2023-10-30T05:33:30Z | 2023-07-24T15:04:08Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | garyhxfang | 84,261,039 | MDQ6VXNlcjg0MjYxMDM5 | User | false |
huggingface/diffusers | 1,586,330,069 | I_kwDOHa8MBc5ejXnV | 2,364 | https://github.com/huggingface/diffusers/issues/2364 | https://api.github.com/repos/huggingface/diffusers/issues/2364 | Unable to run training script | ### Describe the bug
I am trying to train my model following the instruction here - https://github.com/huggingface/diffusers/tree/main/examples/dreambooth#training-on-a-12gb-gpu.
I am using a SageMaker Notebook instance running on a ml.g4dn.xlarge image. I have full internet access from my notebook. Can I get some ... | closed | completed | false | 0 | [
"bug"
] | [] | 2023-02-15T18:18:03Z | 2023-02-15T19:56:30Z | 2023-02-15T19:56:30Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | kirit93 | 8,190,572 | MDQ6VXNlcjgxOTA1NzI= | User | false |
huggingface/diffusers | 1,586,929,612 | I_kwDOHa8MBc5elp_M | 2,370 | https://github.com/huggingface/diffusers/issues/2370 | https://api.github.com/repos/huggingface/diffusers/issues/2370 | training Lora with 2.1 able on a 24 GB GPU? | Hello!
I really would like to train Lora on my 24 GB GPU but I always get OOM. Is there any parameter or setting which I missed and which could help to reduce the needed VRAM?
```
accelerate launch --mixed_precision="bf16" training.py \
--pretrained_model_name_or_path=$MODEL_NAME \
--dataset_name=$DATASET_... | closed | completed | false | 13 | [] | [] | 2023-02-16T03:11:11Z | 2023-03-21T09:37:54Z | 2023-03-21T02:23:33Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Marcophono2 | 22,599,855 | MDQ6VXNlcjIyNTk5ODU1 | User | false |
huggingface/diffusers | 1,586,941,382 | I_kwDOHa8MBc5els3G | 2,371 | https://github.com/huggingface/diffusers/issues/2371 | https://api.github.com/repos/huggingface/diffusers/issues/2371 | Objects From Dreambooth Training Are Not in Output | ### Describe the bug
I am trying to use the train_dreambooth.py script with the example [images of a dog](https://drive.google.com/drive/folders/1BO_dyz-p65qhBRRMRA4TbZ8qW4rB99JZ).
When I run inferencing on the model that is output from the training, the dog is never in the images being rendered at all. I have tri... | closed | completed | false | 8 | [
"bug"
] | [
"williamberman",
"patrickvonplaten"
] | 2023-02-16T03:27:34Z | 2023-03-16T10:41:59Z | 2023-03-16T10:41:59Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | rajbala | 153,332 | MDQ6VXNlcjE1MzMzMg== | User | false |
huggingface/diffusers | 1,587,098,994 | I_kwDOHa8MBc5emTVy | 2,372 | https://github.com/huggingface/diffusers/issues/2372 | https://api.github.com/repos/huggingface/diffusers/issues/2372 | Generated LoRA doesn't generate expected faces during inference | I generated a LoRA with the following command. My `dataset_name` points to a folder with 17 images of my face, and a `metadata.jsonl` with the appropriate format required by 🤗 datasets. The pretrained model is a merge of two models found on civitai, then converted to the required `diffusers` format.
```shell
acc... | closed | completed | false | 6 | [
"stale"
] | [] | 2023-02-16T06:25:27Z | 2023-03-26T15:02:57Z | 2023-03-26T15:02:57Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | jndietz | 4,515,791 | MDQ6VXNlcjQ1MTU3OTE= | User | false |
huggingface/diffusers | 1,587,231,138 | I_kwDOHa8MBc5emzmi | 2,375 | https://github.com/huggingface/diffusers/issues/2375 | https://api.github.com/repos/huggingface/diffusers/issues/2375 | NameError: name 'safe_open' is not defined (same code running fine yesterday) | ### System Info
Running in Google Colab:
- `transformers` version: 4.26.1
- Platform: Linux-5.10.147+-x86_64-with-glibc2.29
- Python version: 3.8.10
- Huggingface_hub version: 0.12.0
- PyTorch version (GPU?): 1.13.1+cu116 (True)
- Tensorflow version (GPU?): 2.11.0 (True)
- Flax version (CPU?/GPU?/TPU?): not... | closed | completed | false | 7 | [] | [] | 2023-02-16T00:42:10Z | 2024-05-07T12:55:25Z | 2023-02-19T02:41:45Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | VijCodes | 22,949,079 | MDQ6VXNlcjIyOTQ5MDc5 | User | false |
huggingface/diffusers | 1,587,559,411 | I_kwDOHa8MBc5eoDvz | 2,377 | https://github.com/huggingface/diffusers/issues/2377 | https://api.github.com/repos/huggingface/diffusers/issues/2377 | [Utils] Implement tests for `EMAModel` | Reference: https://github.com/huggingface/diffusers/pull/2302#issuecomment-1432873275 | closed | completed | false | 0 | [] | [
"sayakpaul"
] | 2023-02-16T12:06:05Z | 2023-03-14T05:24:47Z | 2023-03-14T05:24:47Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | sayakpaul | 22,957,388 | MDQ6VXNlcjIyOTU3Mzg4 | User | false |
huggingface/diffusers | 1,588,158,256 | I_kwDOHa8MBc5eqV8w | 2,379 | https://github.com/huggingface/diffusers/issues/2379 | https://api.github.com/repos/huggingface/diffusers/issues/2379 | Add support for converting inpainting,instruct-pix2pix,etc models | The conversion scripts can convert ckpt and safetensors models to diffusers format, but it does not work for inpainting or instruct-pix2pix models:
Example: Using this safetensors file: https://huggingface.co/stabilityai/stable-diffusion-2-inpainting/blob/main/512-inpainting-ema.safetensors
`python3 convert_origi... | closed | completed | false | 4 | [] | [] | 2023-02-16T18:20:08Z | 2023-04-12T15:14:09Z | 2023-03-08T21:17:07Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | reimager | 20,918,094 | MDQ6VXNlcjIwOTE4MDk0 | User | false |
huggingface/diffusers | 1,348,051,555 | I_kwDOHa8MBc5QWaJj | 239 | https://github.com/huggingface/diffusers/issues/239 | https://api.github.com/repos/huggingface/diffusers/issues/239 | indices should be either on cpu or on the same device as the indexed tensor (cpu) | ### Describe the bug
image_to_image.py line 92 throws the error above.
` init_latents = self.scheduler.add_noise(init_latents, noise, timesteps)`
I've tried adding .to(self.device) to the 3 parameters.
Device should be 'cuda' though.
### Reproduction
```
device = "cuda"
pipei2i = StableDiff... | closed | completed | false | 21 | [
"bug"
] | [] | 2022-08-23T14:28:10Z | 2022-09-13T16:15:44Z | 2022-09-13T16:15:44Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Vortitron | 8,288,276 | MDQ6VXNlcjgyODgyNzY= | User | false |
huggingface/diffusers | 1,588,792,058 | I_kwDOHa8MBc5eswr6 | 2,390 | https://github.com/huggingface/diffusers/issues/2390 | https://api.github.com/repos/huggingface/diffusers/issues/2390 | [Pipelines] Support for T2I-Adapter | ### Model/Pipeline/Scheduler description
From the official repository, [T2I-Adapter](https://github.com/TencentARC/T2I-Adapter) by @TencentARC is
> ... a simple and small (~70M parameters, ~300M storage space) network that can provide extra guidance to pre-trained text-to-image models while freezing the original la... | closed | completed | false | 12 | [
"stale"
] | [] | 2023-02-17T05:49:35Z | 2023-04-13T15:03:34Z | 2023-04-13T15:03:34Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | wfng92 | 43,742,196 | MDQ6VXNlcjQzNzQyMTk2 | User | false |
huggingface/diffusers | 1,588,870,571 | I_kwDOHa8MBc5etD2r | 2,392 | https://github.com/huggingface/diffusers/issues/2392 | https://api.github.com/repos/huggingface/diffusers/issues/2392 | Using StableDiffusionImg2ImgPipeline with prompt_embeds returns ValueError: `prompt` has to be of type `str` or `list` but is <class 'NoneType'> | ### Describe the bug
I'm trying to do inference with `StableDiffusionImg2ImgPipeline` using prompt embeddings rather than prompt as a string. I am using the [Compel](https://github.com/damian0815/compel) library to create the embeddings. This works well with `StableDiffusionPipeline` but returns the following error ... | closed | completed | false | 6 | [
"bug"
] | [] | 2023-02-17T07:20:48Z | 2023-03-06T11:27:01Z | 2023-02-21T06:03:57Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | alexisrolland | 13,064,696 | MDQ6VXNlcjEzMDY0Njk2 | User | false |
huggingface/diffusers | 1,588,933,250 | I_kwDOHa8MBc5etTKC | 2,394 | https://github.com/huggingface/diffusers/issues/2394 | https://api.github.com/repos/huggingface/diffusers/issues/2394 | How to remove Logs， | I want to use pyinstaller to package as exe without cmd command line
| closed | completed | false | 4 | [
"stale"
] | [] | 2023-02-17T08:24:14Z | 2023-04-14T15:03:30Z | 2023-04-14T15:03:30Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | zhanghongyong123456 | 48,466,610 | MDQ6VXNlcjQ4NDY2NjEw | User | false |
huggingface/diffusers | 1,589,443,352 | I_kwDOHa8MBc5evPsY | 2,399 | https://github.com/huggingface/diffusers/issues/2399 | https://api.github.com/repos/huggingface/diffusers/issues/2399 | [Tests] Add tests for DDIM Inversion | @patrickvonplaten reminder! | closed | completed | false | 1 | [] | [
"patrickvonplaten"
] | 2023-02-17T14:22:14Z | 2023-03-29T15:12:24Z | 2023-03-29T15:12:24Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | patrickvonplaten | 23,423,619 | MDQ6VXNlcjIzNDIzNjE5 | User | false |
huggingface/diffusers | 1,275,866,666 | I_kwDOHa8MBc5MDC4q | 24 | https://github.com/huggingface/diffusers/issues/24 | https://api.github.com/repos/huggingface/diffusers/issues/24 | Friction log from new user | Hey all! I'm writing a friction log as a new user and someone with 0 experience with diffusion models. I hope this is useful.
Overall, I **loved** that the README had a definition section, it was very quick to get things going and I was able to have some result with existing pipelines very quickly. As a first step I... | closed | completed | false | 4 | [] | [] | 2022-06-18T19:29:45Z | 2022-07-21T18:56:14Z | 2022-07-21T18:56:14Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | osanseviero | 7,246,357 | MDQ6VXNlcjcyNDYzNTc= | User | false |
huggingface/diffusers | 1,589,671,911 | I_kwDOHa8MBc5ewHfn | 2,402 | https://github.com/huggingface/diffusers/issues/2402 | https://api.github.com/repos/huggingface/diffusers/issues/2402 | AssertionError: Torch not compiled with CUDA enabled | ### Describe the bug
I followed the guidelines in the README.md file and tried to run the pipeline with CUDA enabled. I am getting the following error:
```python
python main.py
C:\Users\notdanilo\AppData\Local\Programs\Python\Python310\lib\site-packages\transformers\models\clip\feature_extraction_clip.py:28: Futu... | closed | completed | false | 1 | [
"bug"
] | [] | 2023-02-17T16:52:50Z | 2023-02-17T22:29:47Z | 2023-02-17T22:29:46Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | notdanilo | 1,069,140 | MDQ6VXNlcjEwNjkxNDA= | User | false |
huggingface/diffusers | 1,590,168,625 | I_kwDOHa8MBc5eyAwx | 2,410 | https://github.com/huggingface/diffusers/issues/2410 | https://api.github.com/repos/huggingface/diffusers/issues/2410 | train_text_to_image_flax.py no flax_model.msgpack or pytorch_model.bin | ### Describe the bug
Hi, I'm in colab pro environment using TPU v2 for the test purpose.
I get this error
```OSError: CompVis/stable-diffusion-v1-4 does not appear to have a file named
flax_model.msgpack or pytorch_model.bin.
```
```
2023-02-18 02:07:26.213102: W tensorflow/compiler/xla/stream_executor/pl... | closed | completed | false | 2 | [
"bug"
] | [
"pcuenca"
] | 2023-02-18T02:20:56Z | 2023-03-06T19:46:23Z | 2023-03-06T19:46:22Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | treksis | 54,779,707 | MDQ6VXNlcjU0Nzc5NzA3 | User | false |
huggingface/diffusers | 1,590,226,080 | I_kwDOHa8MBc5eyOyg | 2,411 | https://github.com/huggingface/diffusers/issues/2411 | https://api.github.com/repos/huggingface/diffusers/issues/2411 | pre-trained stable diffusion model taking 4min to generate an image | hi , I'm trying to generate an image from a prompt using the Stable-Diffusion-v1-5 model,and I'm using NVidia GeForce GTX 1650 , to create an image it's taking 4 min, is it normal? or I did something wrong?, how can I improve the performance | closed | completed | false | 4 | [] | [] | 2023-02-18T06:08:16Z | 2023-02-21T08:20:30Z | 2023-02-21T08:20:03Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | HEGADE | 60,566,946 | MDQ6VXNlcjYwNTY2OTQ2 | User | false |
huggingface/diffusers | 1,590,347,584 | I_kwDOHa8MBc5eysdA | 2,412 | https://github.com/huggingface/diffusers/issues/2412 | https://api.github.com/repos/huggingface/diffusers/issues/2412 | Error using dreambooth | ### Describe the bug
Following the tutorial in notebook:
https://github.com/huggingface/diffusers/tree/main/examples/dreambooth
with latest version of code I have follow error:
ImportError: cannot import name 'ProjectConfiguration' from 'accelerate.utils'
With a previous commit (i.e. 1e7f96544203724652707dc93... | closed | completed | false | 2 | [
"bug"
] | [] | 2023-02-18T13:35:24Z | 2023-02-20T22:36:50Z | 2023-02-20T22:36:49Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | EnricoBeltramo | 22,332,651 | MDQ6VXNlcjIyMzMyNjUx | User | false |
huggingface/diffusers | 1,590,369,392 | I_kwDOHa8MBc5eyxxw | 2,413 | https://github.com/huggingface/diffusers/issues/2413 | https://api.github.com/repos/huggingface/diffusers/issues/2413 | stabilityai/stable-diffusion-2 is missing fp16 files | null | closed | completed | false | 1 | [] | [] | 2023-02-18T14:57:45Z | 2023-02-21T08:27:33Z | 2023-02-21T08:27:33Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | antgame11 | 68,531,196 | MDQ6VXNlcjY4NTMxMTk2 | User | false |
huggingface/diffusers | 1,590,463,381 | I_kwDOHa8MBc5ezIuV | 2,414 | https://github.com/huggingface/diffusers/issues/2414 | https://api.github.com/repos/huggingface/diffusers/issues/2414 | "NameError: name 'transformers' is not defined" when i load model from hub | ### Describe the bug
An error occurs when trying to load any model from the hub:
```
[/usr/local/lib/python3.8/dist-packages/diffusers/pipelines/pipeline_utils.py](https://localhost:8080/#) in from_pretrained(cls, pretrained_model_name_or_path, **kwargs)
826
827 is_diffusers_model = issubc... | closed | completed | false | 3 | [
"bug"
] | [] | 2023-02-18T20:37:01Z | 2023-02-20T07:34:14Z | 2023-02-20T07:34:14Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | WiNE-iNEFF | 41,611,046 | MDQ6VXNlcjQxNjExMDQ2 | User | false |
huggingface/diffusers | 1,590,486,508 | I_kwDOHa8MBc5ezOXs | 2,416 | https://github.com/huggingface/diffusers/issues/2416 | https://api.github.com/repos/huggingface/diffusers/issues/2416 | Train text to image slower with xformers | ### Describe the bug
The train text to image is slower when xformers 0.0.16 is enabled (6 hours and 4 mins) vs when x formers is not used (5 hours and 45 mins) on a A100 40GB GPU. About the same amount of memory (20 GB) is used in each case.
### Reproduction
Comment out uunet.enable_xformers_memory_efficient_attenti... | closed | completed | false | 4 | [
"bug",
"stale"
] | [] | 2023-02-18T22:16:59Z | 2023-03-29T15:12:22Z | 2023-03-29T15:12:22Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | gltanaka | 24,632,716 | MDQ6VXNlcjI0NjMyNzE2 | User | false |
huggingface/diffusers | 1,590,637,798 | I_kwDOHa8MBc5ezzTm | 2,419 | https://github.com/huggingface/diffusers/issues/2419 | https://api.github.com/repos/huggingface/diffusers/issues/2419 | CompVis/stable-diffusion-v1-4 is missing fp16 files | *Problem*
You are loading the variant fp16 from CompVis/stable-diffusion-v1-4 via `revision='fp16'`. This behavior is deprecated and will be removed in diffusers v1. One should use `variant='fp16'` instead. However, it appears that CompVis/stable-diffusion-v1-4 currently does not have the required variant filenames ... | closed | completed | false | 2 | [] | [] | 2023-02-19T10:31:48Z | 2023-02-20T07:50:42Z | 2023-02-20T07:43:12Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | vilobhmm | 6,590,568 | MDQ6VXNlcjY1OTA1Njg= | User | false |
huggingface/diffusers | 1,590,670,991 | I_kwDOHa8MBc5ez7aP | 2,420 | https://github.com/huggingface/diffusers/issues/2420 | https://api.github.com/repos/huggingface/diffusers/issues/2420 | train_text_to_image_lora.py faces a problem when enable_xformers_memory_efficient_attention is True | ### Describe the bug
Hello,
When I fine tune the model using this [code](https://github.com/huggingface/diffusers/blob/main/examples/text_to_image/train_text_to_image_lora.py) and I enable enable_xformers_memory_efficient_attention, I face the following bug:
```
accelerator.backward(loss)
File "C:\Users\En... | closed | completed | false | 5 | [
"bug",
"stale"
] | [] | 2023-02-19T12:31:35Z | 2023-03-29T15:12:21Z | 2023-03-29T15:12:21Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | FBehrad | 41,340,554 | MDQ6VXNlcjQxMzQwNTU0 | User | false |
huggingface/diffusers | 1,590,831,533 | I_kwDOHa8MBc5e0imt | 2,424 | https://github.com/huggingface/diffusers/issues/2424 | https://api.github.com/repos/huggingface/diffusers/issues/2424 | Enable flake8-comprehension checks for better / faster code | **Is your feature request related to a problem? Please describe.**
Enable flake8-comprehensions which is particularly trivial since the ruff linter was recently installed. It removes useless or inefficient Python comprehensions which should marginally improve performance and improve readability.
**Describe the solu... | closed | completed | false | 2 | [
"stale"
] | [] | 2023-02-19T19:55:09Z | 2023-03-27T14:19:33Z | 2023-03-27T14:19:32Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Skylion007 | 2,053,727 | MDQ6VXNlcjIwNTM3Mjc= | User | false |
huggingface/diffusers | 1,590,909,844 | I_kwDOHa8MBc5e01uU | 2,427 | https://github.com/huggingface/diffusers/issues/2427 | https://api.github.com/repos/huggingface/diffusers/issues/2427 | Muse | ### Model/Pipeline/Scheduler description
We are working on the training script for [Muse](https://muse-model.github.io/) by google, the progress can be seen [here](https://github.com/lucidrains/muse-maskgit-pytorch/pull/25).
I'm trying to make it in the huggingface format as much as I can so that when it's done, th... | closed | completed | false | 2 | [
"stale"
] | [] | 2023-02-19T23:45:01Z | 2023-04-07T15:03:37Z | 2023-04-07T15:03:37Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | isamu-isozaki | 23,430,101 | MDQ6VXNlcjIzNDMwMTAx | User | false |
huggingface/diffusers | 1,591,308,292 | I_kwDOHa8MBc5e2XAE | 2,429 | https://github.com/huggingface/diffusers/issues/2429 | https://api.github.com/repos/huggingface/diffusers/issues/2429 | Is it good to train on 512x512, inference on 512x768? | Thank you for your great job. It is really useful for me. I want to inference with resolution of 512x768. I know I could do that with the model trained on 512x512. But to get the best performance on both 512x512 and 512x768, should I trained on 512x512 or 512x768 or 768x768. I'd appreciate your advise. | closed | completed | false | 10 | [
"stale"
] | [] | 2023-02-20T07:30:39Z | 2023-04-16T15:03:25Z | 2023-04-16T15:03:25Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | tengshaofeng | 9,246,739 | MDQ6VXNlcjkyNDY3Mzk= | User | false |
huggingface/diffusers | 1,591,392,240 | I_kwDOHa8MBc5e2rfw | 2,431 | https://github.com/huggingface/diffusers/issues/2431 | https://api.github.com/repos/huggingface/diffusers/issues/2431 | Is it possible diffusers implement an official support on the increasing or decreasing weight of prompt with () & []? | **Is your feature request related to a problem? Please describe.**
The currently the AUTOMATIC1111/stable-diffusion-web-ui support to increase or decrease the weight of an prompt with () & [] which is not supported by diffusers.
(e.g. "best_quality (1girl:1.3) bow bride brown_hair closed_mouth frilled_bow frilled_h... | closed | completed | false | 51 | [
"stale"
] | [
"patrickvonplaten"
] | 2023-02-20T08:37:26Z | 2023-10-09T06:46:50Z | 2023-08-16T15:05:04Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | garyhxfang | 84,261,039 | MDQ6VXNlcjg0MjYxMDM5 | User | false |
huggingface/diffusers | 1,591,535,884 | I_kwDOHa8MBc5e3OkM | 2,432 | https://github.com/huggingface/diffusers/issues/2432 | https://api.github.com/repos/huggingface/diffusers/issues/2432 | [Pipeline] Extending Stable Diffusion for generating videos | Since [Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation](https://arxiv.org/abs/2212.11565) has been there for sometime now, it'd be cool to officially have it supported from Diffusers 🧨
The best part is the official repository (https://github.com/showlab/Tune-A-Video) itself bui... | closed | completed | false | 19 | [
"stale"
] | [] | 2023-02-20T10:08:52Z | 2023-04-20T15:03:38Z | 2023-04-20T15:03:38Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | sayakpaul | 22,957,388 | MDQ6VXNlcjIyOTU3Mzg4 | User | false |
huggingface/diffusers | 1,591,911,272 | I_kwDOHa8MBc5e4qNo | 2,435 | https://github.com/huggingface/diffusers/issues/2435 | https://api.github.com/repos/huggingface/diffusers/issues/2435 | A bug in multiple gpu training | ### Describe the bug
when train this code with multiple gpu
I got a bug like this
![截屏2023-02-20 下午9 38 48](https://user-images.githubusercontent.com/117744200/220123555-923071d6-984e-48... | closed | completed | false | 2 | [
"bug",
"stale"
] | [] | 2023-02-20T13:52:29Z | 2023-04-07T15:03:31Z | 2023-04-07T15:03:31Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | kangxi1818 | 117,744,200 | U_kgDOBwSiSA | User | false |
huggingface/diffusers | 1,592,361,772 | I_kwDOHa8MBc5e6YMs | 2,437 | https://github.com/huggingface/diffusers/issues/2437 | https://api.github.com/repos/huggingface/diffusers/issues/2437 | [Pipelines] Add a T2I-Adapter pipeline | ### Model/Pipeline/Scheduler description
T2I-Adapter is a simple and small (~70M parameters, ~300M storage space) network that can provide extra guidance to pre-trained text-to-image models while freezing the original large text-to-image models.
It has [3 pre-trained models ](https://huggingface.co/TencentARC/T2I-A... | closed | completed | false | 2 | [
"stale"
] | [] | 2023-02-20T19:38:34Z | 2023-03-31T15:03:18Z | 2023-03-31T15:03:18Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | huo-ju | 561,810 | MDQ6VXNlcjU2MTgxMA== | User | false |
huggingface/diffusers | 1,592,403,149 | I_kwDOHa8MBc5e6iTN | 2,438 | https://github.com/huggingface/diffusers/issues/2438 | https://api.github.com/repos/huggingface/diffusers/issues/2438 | About the fourier time embedding type | Dear Develops,
I noticed that in
https://github.com/huggingface/diffusers/blob/17ecf72d4472f5dc11e7c86841c95898f6edbc0b/src/diffusers/models/unet_2d.py#L307-#L309, you scaled the samples with the timestep when the time embedding type is fourier. But you did not apply such normalization for the unet2d_condition an... | closed | completed | false | 4 | [] | [] | 2023-02-20T20:24:48Z | 2023-03-13T12:16:23Z | 2023-03-13T12:16:23Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | WHan-alter | 72,044,555 | MDQ6VXNlcjcyMDQ0NTU1 | User | false |
huggingface/diffusers | 1,592,547,891 | I_kwDOHa8MBc5e7Foz | 2,439 | https://github.com/huggingface/diffusers/issues/2439 | https://api.github.com/repos/huggingface/diffusers/issues/2439 | Issue loading and running inpainting pipeline | ### Describe the bug
Following this tutorial:
https://huggingface.co/runwayml/stable-diffusion-inpainting
I tried to made an inference with and image jpg of size 283x530 (and a mask png of same size) , but I have an error:
Sizes of tensors must match except in dimension 1. Expected size 64 but got size 66 for t... | closed | completed | false | 1 | [
"bug"
] | [] | 2023-02-20T23:07:08Z | 2023-02-21T21:44:33Z | 2023-02-21T21:44:33Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | EnricoBeltramo | 22,332,651 | MDQ6VXNlcjIyMzMyNjUx | User | false |
huggingface/diffusers | 1,349,412,333 | I_kwDOHa8MBc5QbmXt | 244 | https://github.com/huggingface/diffusers/issues/244 | https://api.github.com/repos/huggingface/diffusers/issues/244 | Add information for contributors on how to apply code formatter | **Is your feature request related to a problem? Please describe.**
I don't know how we are supposed to apply code formatters. I found settings for `black`, `isort` and `flake8` in `pyproject.toml` and `setup.cfg`, but it doesn't seem to be used. When I apply `isort` to the `main` branch, it changes codes. I think co... | closed | completed | false | 4 | [] | [
"anton-l"
] | 2022-08-24T13:01:42Z | 2022-09-13T15:55:41Z | 2022-09-13T15:55:40Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | hysts | 25,161,192 | MDQ6VXNlcjI1MTYxMTky | User | false |
huggingface/diffusers | 1,592,552,770 | I_kwDOHa8MBc5e7G1C | 2,440 | https://github.com/huggingface/diffusers/issues/2440 | https://api.github.com/repos/huggingface/diffusers/issues/2440 | Error load inpainting model in float16 | ### Describe the bug
Following this tutorial:
https://huggingface.co/runwayml/stable-diffusion-inpainting
I tried to made to load the model with float16, but I have follow error:
RuntimeError: "LayerNormKernelImpl" not implemented for 'Half'
If I try without float16, the model load correctly
### Reproduction
... | closed | completed | false | 2 | [
"bug"
] | [] | 2023-02-20T23:12:58Z | 2023-02-25T08:56:18Z | 2023-02-25T08:56:18Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | EnricoBeltramo | 22,332,651 | MDQ6VXNlcjIyMzMyNjUx | User | false |
huggingface/diffusers | 1,592,597,049 | I_kwDOHa8MBc5e7Ro5 | 2,441 | https://github.com/huggingface/diffusers/issues/2441 | https://api.github.com/repos/huggingface/diffusers/issues/2441 | stabilityai/stable-diffusion-2-1 is missing fp16 files | diffusers told me i should post this | closed | completed | false | 1 | [] | [] | 2023-02-21T00:22:19Z | 2023-02-21T08:25:53Z | 2023-02-21T08:25:53Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | damian0815 | 144,366 | MDQ6VXNlcjE0NDM2Ng== | User | false |
huggingface/diffusers | 1,592,744,687 | I_kwDOHa8MBc5e71rv | 2,442 | https://github.com/huggingface/diffusers/issues/2442 | https://api.github.com/repos/huggingface/diffusers/issues/2442 | How should the loss graph look when finetuning an inpainting model? | @thedarkzeno @patil-suraj thank you for the amazing work here!!!: https://github.com/huggingface/diffusers/tree/main/examples/research_projects/dreambooth_inpaint
**Questions:**
1. Should loss be decreasing while running train_dreambooth_inpaint.py?
2. If it should, how would you go about debugging why it isn't?
... | closed | completed | false | 6 | [] | [] | 2023-02-21T04:06:28Z | 2023-03-01T05:34:40Z | 2023-03-01T05:34:40Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | lakshman111 | 4,174,501 | MDQ6VXNlcjQxNzQ1MDE= | User | false |
huggingface/diffusers | 1,592,811,837 | I_kwDOHa8MBc5e8GE9 | 2,443 | https://github.com/huggingface/diffusers/issues/2443 | https://api.github.com/repos/huggingface/diffusers/issues/2443 | [SD2.1] - Input shapes for unet model | Trying to run via Openvino IR - inferencing a pixelated image currently
Input Nodes for SD2.1:
sample - [2,4,64,64],timestep [-1] and encoder_hidden_states [2,77,1024]
Still I do get the inferenced image as 512x512 since vae_decoder takes latents input of shape 512x512 and that results in a pixelated image. Wha... | closed | completed | false | 9 | [
"stale"
] | [] | 2023-02-21T05:08:55Z | 2023-04-15T15:03:42Z | 2023-04-15T15:03:42Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | lalith-mcw | 84,698,958 | MDQ6VXNlcjg0Njk4OTU4 | User | false |
huggingface/diffusers | 1,592,879,389 | I_kwDOHa8MBc5e8Wkd | 2,444 | https://github.com/huggingface/diffusers/issues/2444 | https://api.github.com/repos/huggingface/diffusers/issues/2444 | train_dreambooth_inpaint.py throwing "returned non-zero exit status 1" error | ### Describe the bug
@thedarkzeno @patil-suraj
I've been running train_dreambooth_inpaint.py from https://github.com/huggingface/diffusers/tree/main/examples/research_projects/dreambooth_inpaint for the last few days with the same environment configurations. Today I ran it again and got:
Traceback (most rece... | closed | completed | false | 7 | [
"bug",
"stale"
] | [] | 2023-02-21T06:16:47Z | 2023-04-15T15:03:41Z | 2023-04-15T15:03:41Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | lakshman111 | 4,174,501 | MDQ6VXNlcjQxNzQ1MDE= | User | false |
huggingface/diffusers | 1,593,036,129 | I_kwDOHa8MBc5e881h | 2,445 | https://github.com/huggingface/diffusers/issues/2445 | https://api.github.com/repos/huggingface/diffusers/issues/2445 | Load safetensors directly to cuda | As far as I know there is not way right now to load a model from a safetensors file directly to cuda. You always have to load it to cpu first. Safetensors library supports loading directly to cuda, so it shouldn't be hard to add this functionality to diffusers pipelines.
Interface may look like this (just specify de... | closed | completed | false | 9 | [] | [] | 2023-02-21T08:25:16Z | 2023-03-09T12:06:37Z | 2023-03-08T12:14:11Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Daniel-Kelvich | 19,155,101 | MDQ6VXNlcjE5MTU1MTAx | User | false |
huggingface/diffusers | 1,593,109,705 | I_kwDOHa8MBc5e9OzJ | 2,447 | https://github.com/huggingface/diffusers/issues/2447 | https://api.github.com/repos/huggingface/diffusers/issues/2447 | LoRA error | Hi,
I was experiencing this error frequently 4 among 10 trainings. Has anyone faced or know about this on how to get around?
CC: @patrickvonplaten @patil-suraj @cloneofsimo
02/19/2023 10:40:23 - INFO - __main__ - ***** Running training *****
02/19/2023 10:40:23 - INFO - __main__ - Num examples = 4
02/19/2... | closed | completed | false | 7 | [
"stale"
] | [] | 2023-02-21T09:14:00Z | 2023-05-01T15:03:36Z | 2023-05-01T15:03:36Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | polavishnu4444 | 2,263,240 | MDQ6VXNlcjIyNjMyNDA= | User | false |
huggingface/diffusers | 1,594,038,274 | I_kwDOHa8MBc5fAxgC | 2,453 | https://github.com/huggingface/diffusers/issues/2453 | https://api.github.com/repos/huggingface/diffusers/issues/2453 | How can I contribute? | **Is your feature request related to a problem? Please describe.**
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
**Describe the solution you'd like**
A clear and concise description of what you want to happen.
**Describe alternatives you've considered**
A clear and conci... | closed | completed | false | 5 | [
"stale"
] | [] | 2023-02-21T20:09:29Z | 2023-04-26T23:28:23Z | 2023-04-26T23:28:23Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | qwapilot | 111,024,508 | U_kgDOBp4ZfA | User | false |
huggingface/diffusers | 1,594,275,086 | I_kwDOHa8MBc5fBrUO | 2,454 | https://github.com/huggingface/diffusers/issues/2454 | https://api.github.com/repos/huggingface/diffusers/issues/2454 | Loading the model from a single *.safetensors file in the root of a repo | Hello, I'm scrutinizing the docs and PRs and cannot find an answer.
Is there a way to do `from_pretrained` on the folder which contains a single `*.safetensors` file with a json config?
Or it has to be a set of folders containing each part in safetensors format?
| closed | completed | false | 14 | [] | [] | 2023-02-22T00:36:50Z | 2023-03-17T09:21:58Z | 2023-02-23T16:57:58Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | zetyquickly | 25,350,960 | MDQ6VXNlcjI1MzUwOTYw | User | false |
huggingface/diffusers | 1,595,002,319 | I_kwDOHa8MBc5fEc3P | 2,457 | https://github.com/huggingface/diffusers/issues/2457 | https://api.github.com/repos/huggingface/diffusers/issues/2457 | img2img style transfer | For style transfer using the img2img function, does the input image need to be resized?
For example, using sd-2-1-base, any image size input will always produce completely unusable results.
Have any suggestions? Thanks | closed | completed | false | 5 | [
"stale"
] | [] | 2023-02-22T12:14:48Z | 2023-04-01T15:03:20Z | 2023-04-01T15:03:20Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Jason-kid | 31,425,629 | MDQ6VXNlcjMxNDI1NjI5 | User | false |
huggingface/diffusers | 1,595,045,349 | I_kwDOHa8MBc5fEnXl | 2,459 | https://github.com/huggingface/diffusers/issues/2459 | https://api.github.com/repos/huggingface/diffusers/issues/2459 | “unet.enable_xformers_memory_efficient_attention()” doesn't work in train_text_to_image_lora.py | ### Describe the bug
`unet.enable_xformers_memory_efficient_attention() ` should be placed after `unet.set_attn_processor(lora_attn_procs) `, otherwise lora's weight will not use xformers.
I can use `enable_xformers_memory_efficient_attention` correctly after this modification:
```
unet.set_attn_processor(l... | closed | completed | false | 3 | [
"bug"
] | [] | 2023-02-22T12:46:04Z | 2024-01-15T19:08:45Z | 2023-02-23T05:38:23Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | ZihaoW123 | 47,708,655 | MDQ6VXNlcjQ3NzA4NjU1 | User | false |
huggingface/diffusers | 1,350,060,851 | I_kwDOHa8MBc5QeEsz | 246 | https://github.com/huggingface/diffusers/issues/246 | https://api.github.com/repos/huggingface/diffusers/issues/246 | AttributeError: module transformers has no attribute FeatureExtractionMixin | ### Describe the bug
Getting the error
```
AttributeError: module transformers has no attribute FeatureExtractionMixin
```
when running the example script provided for StableDiffusio v1.3:
https://huggingface.co/CompVis/stable-diffusion-v1-3
### Reproduction
```python
# make sure you're logged in wit... | closed | completed | false | 2 | [
"bug"
] | [] | 2022-08-24T21:51:15Z | 2022-08-25T13:14:29Z | 2022-08-25T13:14:28Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | loretoparisi | 163,333 | MDQ6VXNlcjE2MzMzMw== | User | false |
huggingface/diffusers | 1,595,053,616 | I_kwDOHa8MBc5fEpYw | 2,460 | https://github.com/huggingface/diffusers/issues/2460 | https://api.github.com/repos/huggingface/diffusers/issues/2460 | Code quality check fails even after running 'make style' and 'make quality'. | I was making a PR and ran `make style`and `make quality`, resolved all issues but CI still tells me something is wrong with the new script I'm trying to contribute. | closed | completed | false | 3 | [
"stale"
] | [] | 2023-02-22T12:52:15Z | 2023-04-01T15:03:18Z | 2023-04-01T15:03:18Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | lukovnikov | 1,732,910 | MDQ6VXNlcjE3MzI5MTA= | User | false |
huggingface/diffusers | 1,595,520,724 | I_kwDOHa8MBc5fGbbU | 2,462 | https://github.com/huggingface/diffusers/issues/2462 | https://api.github.com/repos/huggingface/diffusers/issues/2462 | Stable diffusion finetuning support for IterableDataset | **Is your feature request related to a problem? Please describe.**
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
For `examples/text_to_image/train_text_to_image.py`, I am trying to load a `webdataset` which is an `IterableDataset` that doesn't have the `__len__` proper... | closed | completed | false | 2 | [
"stale"
] | [] | 2023-02-22T17:36:07Z | 2023-04-03T15:03:09Z | 2023-04-03T15:03:09Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | samar-khanna | 39,067,298 | MDQ6VXNlcjM5MDY3Mjk4 | User | false |
huggingface/diffusers | 1,596,047,285 | I_kwDOHa8MBc5fIb-1 | 2,465 | https://github.com/huggingface/diffusers/issues/2465 | https://api.github.com/repos/huggingface/diffusers/issues/2465 | Is there a reproducible dreambooth training recipe for this blog? | I see the examples in [blog](https://huggingface.co/blog/dreambooth) generating very solid examples of dreambooth. But with the dog [example](https://drive.google.com/drive/folders/1BO_dyz-p65qhBRRMRA4TbZ8qW4rB99JZ), I tried to train for 800 steps but after 200 steps, whatever prompts I use gives me something like the ... | closed | completed | false | 8 | [
"stale"
] | [] | 2023-02-23T00:44:06Z | 2023-04-03T15:03:08Z | 2023-04-03T15:03:08Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Victor49152 | 111,467,530 | U_kgDOBqTcCg | User | false |
huggingface/diffusers | 1,596,119,152 | I_kwDOHa8MBc5fIthw | 2,466 | https://github.com/huggingface/diffusers/issues/2466 | https://api.github.com/repos/huggingface/diffusers/issues/2466 | [Examples] `checkpoints_total_limit` seems to be not limiting | ### Describe the bug
We introduced `checkpoints_total_limit` in the examples in https://github.com/huggingface/diffusers/pull/2374.
Setting it, however, doesn't limit the number of checkpoints being preserved. Is that expected?
### Reproduction
```bash
accelerate launch examples/text_to_image/train_text_to_ima... | closed | completed | false | 17 | [
"bug",
"stale"
] | [
"williamberman"
] | 2023-02-23T02:33:51Z | 2023-06-15T22:39:04Z | 2023-06-15T22:39:04Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | sayakpaul | 22,957,388 | MDQ6VXNlcjIyOTU3Mzg4 | User | false |
huggingface/diffusers | 1,596,522,524 | I_kwDOHa8MBc5fKQAc | 2,468 | https://github.com/huggingface/diffusers/issues/2468 | https://api.github.com/repos/huggingface/diffusers/issues/2468 | Failing test of `examples/train_unconditional.py` | Reference: https://github.com/huggingface/diffusers/actions/runs/4245740195/jobs/7381643637
I could verify that this happens when installing `accelerate` from `main`, but not in the current release. I'll investigate if it's a regression or an API change. | closed | completed | false | 0 | [] | [] | 2023-02-23T09:40:44Z | 2023-02-24T22:31:18Z | 2023-02-24T22:31:18Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | pcuenca | 1,177,582 | MDQ6VXNlcjExNzc1ODI= | User | false |
huggingface/diffusers | 1,596,859,180 | I_kwDOHa8MBc5fLiMs | 2,469 | https://github.com/huggingface/diffusers/issues/2469 | https://api.github.com/repos/huggingface/diffusers/issues/2469 | LoRA can not be used in text encoder in "train_text_to_image_lora.py"? | It can be seen from [(https://github.com/cloneofsimo/lora)] that fine-tuning the lora for the text encoder works better than the lora for unet. When will this repo support the lora of the text encoder?
| closed | completed | false | 5 | [
"wip"
] | [] | 2023-02-23T13:22:04Z | 2024-11-21T15:37:36Z | 2024-11-21T15:37:35Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | ZihaoW123 | 47,708,655 | MDQ6VXNlcjQ3NzA4NjU1 | User | false |
huggingface/diffusers | 1,597,325,224 | I_kwDOHa8MBc5fNT-o | 2,471 | https://github.com/huggingface/diffusers/issues/2471 | https://api.github.com/repos/huggingface/diffusers/issues/2471 | Attend-and-excite pipeline doesn't work with a size different from the default | ### Describe the bug
The title says all. Any image size different from the default 512x512 (smaller or larger) results in an error.
I don't have time to track down the issue or debug any further, but I thought I'd let you guys know about it.
### Reproduction
Just take the example at https://huggingface.co/docs/di... | closed | completed | false | 3 | [
"bug"
] | [
"yiyixuxu"
] | 2023-02-23T18:17:34Z | 2023-03-06T18:16:40Z | 2023-03-06T18:16:40Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | jorgemcgomes | 3,987,574 | MDQ6VXNlcjM5ODc1NzQ= | User | false |
huggingface/diffusers | 1,350,666,902 | I_kwDOHa8MBc5QgYqW | 248 | https://github.com/huggingface/diffusers/issues/248 | https://api.github.com/repos/huggingface/diffusers/issues/248 | fix off by one in the mask pyramids | null | closed | completed | false | 0 | [] | [] | 2022-08-25T10:16:20Z | 2022-08-25T10:16:32Z | 2022-08-25T10:16:32Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | blefaudeux | 3,072,861 | MDQ6VXNlcjMwNzI4NjE= | User | false |
huggingface/diffusers | 1,598,347,022 | I_kwDOHa8MBc5fRNcO | 2,480 | https://github.com/huggingface/diffusers/issues/2480 | https://api.github.com/repos/huggingface/diffusers/issues/2480 | dreambooth training script dose not make "unet" forler for each checkpointed directory | ### Describe the bug
according to this document, dreambooth training script should make "unet" folder for each checkpointed directory.
but it is not.
https://huggingface.co/docs/diffusers/main/en/training/dreambooth#performing-inference-using-a-saved-checkpoint
### Reproduction
launch dreambooth training scrip... | closed | completed | false | 11 | [
"bug",
"stale"
] | [
"williamberman",
"sayakpaul"
] | 2023-02-24T10:29:13Z | 2023-06-06T15:03:50Z | 2023-06-06T15:03:50Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | yoon28 | 7,078,947 | MDQ6VXNlcjcwNzg5NDc= | User | false |
huggingface/diffusers | 1,598,970,510 | I_kwDOHa8MBc5fTlqO | 2,482 | https://github.com/huggingface/diffusers/issues/2482 | https://api.github.com/repos/huggingface/diffusers/issues/2482 | Training with LoRA: Train loss does not converge | I am running the [`./examples/text_to_image/train_text_to_image.py`](https://github.com/huggingface/diffusers/tree/main/examples/text_to_image#training-with-lora) script on my custom dataset (also tried with the pokemon dataset), it definitely learns something, it gets the domain of the dataset and generate similar ima... | closed | completed | false | 4 | [
"stale"
] | [] | 2023-02-24T16:48:05Z | 2023-04-14T15:03:26Z | 2023-04-14T15:03:26Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | priprapre | 126,275,546 | U_kgDOB4bP2g | User | false |
huggingface/diffusers | 1,599,173,683 | I_kwDOHa8MBc5fUXQz | 2,484 | https://github.com/huggingface/diffusers/issues/2484 | https://api.github.com/repos/huggingface/diffusers/issues/2484 | Add compatible scheduler tests to PipelineTesterMixin | We should add a test that loops over schedulers compatible with the default scheduler for the pipeline and tests that they can all be ran
| closed | completed | false | 1 | [
"stale"
] | [] | 2023-02-24T19:25:53Z | 2023-04-04T15:03:16Z | 2023-04-04T15:03:16Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | williamberman | 6,505,431 | MDQ6VXNlcjY1MDU0MzE= | User | false |
huggingface/diffusers | 1,599,176,822 | I_kwDOHa8MBc5fUYB2 | 2,485 | https://github.com/huggingface/diffusers/issues/2485 | https://api.github.com/repos/huggingface/diffusers/issues/2485 | Add test for enable_vae_slicing to PipelineTesterMixin | null | closed | completed | false | 2 | [
"stale"
] | [] | 2023-02-24T19:29:01Z | 2023-04-05T18:11:37Z | 2023-04-05T18:11:37Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | williamberman | 6,505,431 | MDQ6VXNlcjY1MDU0MzE= | User | false |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.